The Methods For Finding The Symmetry Of The Faces Structure In Human Faces Recognition

The technology of face recognition has recently received significant attention as one of the most active subject in the field of pattern recognition. Which has wide range applications in commercial and law enforcement in the past several years, and then many face recognition conferences appeared, such as AFGR and AVBPA.The technology of the face recognition recognize face from static image and mobile image using computer as auxiliary means. A general statement of the face recognition problem can be formulated as follows: Given photo or video images of a scene, identify or verify one or more persons in the scene using a stored database of faces.The study of face recognition includes three stages: the first stage involves detecting the location of face from background; the second stage requires pretreatment and extraction of pertinent face feature; finally the third stage involves feature matching and recognition. The first two stages are discussed in this dissertation and the main work is faces pretreatment.In this dissertation, first, a particular survey of automatic face detection and recognition was given; then, a method to image pretreatment, to locate human faces and face normalization based on grey image is discussed; last, a rapid algorithm and it's improved algorithm for finding the symmetry of face structure in human faces recognition is presented. This algorithm is based on the character of the complex changes about the gray parameter of human's feature organs, the right positions of the feature organs are decided according to the vertical complexity of the image. Depending on the symmetry feature of human's feature organs, the angle of inclination is calculated out and the face is propped up. The results of experiments proved the validity of this algorithm.